This study focuses on the interaction of individual and family variables in predicting a series of adolescent problem behaviors, e.g. antisocial behavior, academic problems, depression, eating disorders, substance abuse, suicide, and early sexual activity. Multiple dimensions of the family environment will be studied in combination with children's personality variables in models testing the unique and joint effects on problem behaviors. Attempts will be made not only to identify factors that place adolescents at risk for problem behaviors, but to differentiate among discrete problem behaviors based on unique configurations of family and personality influence. The study draws theory and research from psychological, sociological, and clinical disciplines in addressing a number of key limitations to the extant research on adolescent behavior problems. Specifically the proposed project will 1) simultaneously measure a diverse array of problem behaviors, 2) integrate several distinct lines of family theory and research related to problem behaviors: parent-child relations, family functioning, and family stress, 3) test for interactive effects of family environment and child personality, and 4) follow two cohorts of youth, one across the transition to middle school and the other across the transition to high school. The existence of two distinct family environments that place children at risk for problem behaviors are posited, psychological overcontrol and behavioral undercontrol. These are theorized to have differential effects on internalized an externalized problem types. It is also theorized that normative stress (puberty, school transitions) and non-normative stressors (life events) exacerbate the impact of each family type, and that children's personality moderates the relationship between family types and problem behaviors. A proportionate stratified (Hispanic ethnicity) sample of 5th and 8th grade students from the Ogden, Utah City School District will be surveyed in class during the first year assessment. Teachers will also be surveyed at school and students' mothers and fathers will be surveyed by mail. All subjects (students, parents, and teachers) will be assessed by mail once annually over the subsequent three years. The results will be useful for parents, policy makers, educators, researchers and clinicians concerned with the intervention in and prevention of maladaptive behavior.